D
How to calculate calibration error & accuracy of device? What is allowable tolerance? Interested in knowing about gas valve/oil valve & IGV - servo operated. Do we need consider error w.r.to demand or feedback? Valve stroke length ?
If you are referring to the servo-operated fuel control valves and IGVs used on GE-design heavy duty gas turbines, calibration of LVDT feedback should <b>ALWAYS</b> be done with respect to actual, measured position of the device. <b>One is <b>not</b> calibrating the valve or the servo</b>; one is calibrating the position feedback from the LVDTs, which should be as close as possible to the actual position.
How close is "as close as possible"? For GE-design heavy duty gas turbine applications, it should be possible to calibrate fuel control valve LVDT feedback to within 0.5-1.0% of actual position very easily. For the IGVs, it should be possible to calibrate the LVDT feedback to within 0.5 degree of actual position.
When calibrating LVDT feedback, the most important factor is to make sure the feedback accurately reflects the actual position. Once that's done, the feedback is most always very nearly equal to the "demand" (reference; setpoint). If it's not, there can be several issues, including incorrect servo polarity (which should be verified before any LVDT calibration), or incorrect servo current null bias, or incorrect servo gain. But, an acceptable difference between "demand" and feedback (reference and actual) <b>once the LVDT feedback is accurate with respect to actual position</b> is also about 0.5-1.0% for fuel valves, and about 0.25-0.50 degree for IGVs. But, again, this is only after the LVDT feedback is verified to be as accurate as possible with respect to actual device position.
There have been MANY threads on control.com about LVDT calibration. Use the 'Search' function at the far right of the Menu bar at the top of every control.com page.
If you're referring to some other application (other than a GE-design heavy duty gas turbine), disregard all of the above.
How close is "as close as possible"? For GE-design heavy duty gas turbine applications, it should be possible to calibrate fuel control valve LVDT feedback to within 0.5-1.0% of actual position very easily. For the IGVs, it should be possible to calibrate the LVDT feedback to within 0.5 degree of actual position.
When calibrating LVDT feedback, the most important factor is to make sure the feedback accurately reflects the actual position. Once that's done, the feedback is most always very nearly equal to the "demand" (reference; setpoint). If it's not, there can be several issues, including incorrect servo polarity (which should be verified before any LVDT calibration), or incorrect servo current null bias, or incorrect servo gain. But, an acceptable difference between "demand" and feedback (reference and actual) <b>once the LVDT feedback is accurate with respect to actual position</b> is also about 0.5-1.0% for fuel valves, and about 0.25-0.50 degree for IGVs. But, again, this is only after the LVDT feedback is verified to be as accurate as possible with respect to actual device position.
There have been MANY threads on control.com about LVDT calibration. Use the 'Search' function at the far right of the Menu bar at the top of every control.com page.
If you're referring to some other application (other than a GE-design heavy duty gas turbine), disregard all of the above.
D
David
Thank you CSA
You May Also Like
-
ABB, GE, and Schneider Electric Launch Solutions for Industrial Sustainability
by Seth Price
-
Precise Motion With New Servos and Drive Units From Oriental Motor
-
Understanding Industrial Ethernet Connections: RJ45, Fiber, M12, and SFP
-
Cognex Develops 380 Modular Vision Tunnel for Enhanced Barcode Reading
